Photoinduced self-structured surface pattern on a molecular azo glass film: Structure-property relationship and wavelength correlation

Article abstract of DOI:10.1021/la2027253

In this study, three series of star-shaped molecular azo glasses were synthesized, and self-structured surface pattern formation on the azo compound films was studied by laser irradiation at different wavelengths. The molecular azo glasses were synthesized from three core precursors (Tr-AN, Tr-35AN, Tr-H35AN), which were prepared by ring-opening reactions between 1,3,5-triglycidyl isocyanurate and corresponding aniline derivatives. The star-shaped azo compounds were obtained through azo-coupling reactions between the core precursors and diazonium salts of 4-chloroaniline, 4-aminobenzonitrile, and 4-nitroaniline, respectively. By using the two-step reaction scheme, three series of azo compounds with different structures were obtained. The core precursors and azo compounds were characterized by using ¹H NMR, FT-IR, UV-vis, mass spectrometry, and thermal analyses. The self-structured surface pattern formation on films of the azo compounds was studied by irradiating the azo compound films with a normal-incident laser beam at different wavelengths (488, 532, and 589 nm). The results show that the photoinduced surface pattern formation behavior is closely related to the structure of the azo compounds, excitation wavelength, and light polarization conditions. The absorption band position of the π-π* transition is mainly determined by the electron-withdrawing groups on the azo chromophores. When the excitation wavelength is between λ_max and the band tail at the longer wavelength side, the self-structured surface patterns can be more efficiently induced to form on the films. The 3,5-dimethyl substitution on azo chromophores inhibits the surface pattern formation for certain excitation wavelengths. Increasing molecular interaction also shows an effect of restraining the surface pattern formation. The irradiations with linearly and circularly polarized light cause significant differences in the alignment manner of the pillarlike structures and their saturated height.

Full text of DOI:10.1021/la2027253

ARTICLE
pubs.acs.org/Langmuir
Photoinduced Self-Structured Surface Pattern on a Molecular Azo
Glass Film: StructureꢀProperty Relationship and Wavelength
Correlation
Xiaolin Wang, Jianjun Yin, and Xiaogong Wang*
Department of Chemical Engineering, Laboratory for Advanced Materials, Tsinghua University, Beijing 100084,
People's Republic of China
S Supporting Information
b
ABSTRACT: In this study, three series of star-shaped molecular azo
glasses were synthesized, and self-structured surface pattern formation
on the azo compound films was studied by laser irradiation at different
wavelengths. The molecular azo glasses were synthesized from three
core precursors (Tr-AN, Tr-35AN, Tr-H35AN), which were prepared
by ring-opening reactions between 1,3,5-triglycidyl isocyanurate and
corresponding aniline derivatives. The star-shaped azo compounds
were obtained through azo-coupling reactions between the core pre-
cursors and diazonium salts of 4-chloroaniline, 4-aminobenzonitrile,
and 4-nitroaniline, respectively. By using the two-step reaction scheme, three series of azo compounds with different structures were
obtained. The core precursors and azo compounds were characterized by using 1H NMR, FT-IR, UVꢀvis, mass spectrometry, and
thermal analyses. The self-structured surface pattern formation on films of the azo compounds was studied by irradiating the azo
compound films with a normal-incident laser beam at different wavelengths (488, 532, and 589 nm). The results show that the
photoinduced surface pattern formation behavior is closely related to the structure of the azo compounds, excitation wavelength,
and light polarization conditions. The absorption band position of the πꢀπ* transition is mainly determined by the electron-
withdrawing groups on the azo chromophores. When the excitation wavelength is between λmax and the band tail at the longer
wavelength side, the self-structured surface patterns can be more efficiently induced to form on the films. The 3,5-dimethyl
substitution on azo chromophores inhibits the surface pattern formation for certain excitation wavelengths. Increasing molecular
interaction also shows an effect of restraining the surface pattern formation. The irradiations with linearly and circularly polarized
light cause significant differences in the alignment manner of the pillarlike structures and their saturated height.
’ INTRODUCTION
physical process related to photoinduced mass transport. The
mechanism of SRG formation has been investigated to under-
stand the general nature of the process. SRG formation has been
attributed to isomerization-driven free volume expansion in the
bulk,14 the force based on the dipolar interaction with the
optically induced electric field gradient,15 a translational worm-
like diffusion caused by the photoisomerization of the azobenzene
chromophores,16 a mean-field force caused by the molecule
alignment,17 and a photomechanical effect occurring in thin films.18
Recently, another unusual photoinduced surface pattern for-
mation behavior has been reported by Hubert et al.19,20 Instead
of irradiation with the interfering laser beams to inscribe SRGs,
the submicrometer hexagonal patterns on the azo polymer films
were induced by irradiation with a uniform laser beam at normal
incidence.19 The patterns are characterized by regularly spaced
pillarlike structures with a saturated height of about 100 nm. This
effect is known as spontaneous surface pattern formation through
light irradiation.19 The phenomenon is somehow reminiscent of
Photoinduced surface pattern formation of azo polymers
(polymers containing azobenzene and its derivatives) has at-
tracted considerable attention in recent years.1ꢀ4 Photoinduced
surface-relief-grating (SRG) formation on azo polymer films has
been intensively investigated since it was first reported by
Natansohn et al. and Tripathy et al. in 1995.5,6 The study shows
that when exposed to the interfering laser beams, SRGs can be
formed on the azo polymer films at a temperature well below the
glass transition temperatures (Tg) of the polymers. The sinusoid
surface patterns, which are stable at temperatures below the Tg
values of the polymers, can be erased by heating samples to a
temperature above their Tg values. The surface patterns are also
erasable by light irradiation below the Tg values, such as by
irradiation with a uniform circularly polarized laser beam.1,6
Photoinduced SRG formation has been observed for different
types of polymers, such as side-chain azo polymers, main-chain
azo polymers, and hyperbranched azo polymers, among others.1ꢀ7
This novel processing method has been used for applications in
data storage, optical device fabrication, and surface modification.8ꢀ13
The reversible nature suggests that the SRG formation is a
Received: July 15, 2011
Revised:
August 25, 2011
Published: August 29, 2011
r
2011 American Chemical Society
12666
dx.doi.org/10.1021/la2027253 Langmuir 2011, 27, 12666–12676
|

Langmuir
ARTICLE
the laser-induced periodic surface structure (LIPSS), the forma-
tion of parallel ripple structures on polymer surfaces induced by
irradiation with a single beam of an excimer laser or a Nd:YAG
pulsed laser.21ꢀ24 The LIPSS can be observed for many polymers
without a specific requirement for structures and has been
explained by models based on electromagnetic field theory such
as the surface-scattered wave model.25,26 The pattern formation
on azo polymer films makes a clear distinction with the LIPSS by
the self-structured characteristics and lower light intensity used in
the writing process. The influences of the laser intensity, irradia-
tion time, and wavelength have been studied to understand the
self-structured pattern formation on azo polymer films.20,27
The possible application of this effect in highly efficient data
storage has been demonstrated.28 However, in contrast to the
extensive study of SRGs by using different azo polymers, the self-
structured surface pattern formation has been studied only for a
poly(methyl methacrylate)-based copolymer (DR1MA/MMA
35/65), among a few others.20,28
Molecular glass is a new type of glassy material based on well-
designed low-molecular-weight organic compounds.29,30 The
amorphous molecular material shows thermodynamic stability
and glass-transition behavior similar to those of an amorphous
polymer. On the other hand, it possesses well-defined structures
and better reproducible properties in comparison with polymers.
Thin solid films with a smooth surface can be feasibly prepared
from an amorphous molecular material by spin-coating. It has
been reported that SRGs can be inscribed on molecular azo glass
films with a higher formation rate compared to those observed
for azo polymers.31ꢀ34 Very recently, it has been reported that
complex periodic superstructures can be formed on films of
small-molecule-based photochromic materials upon irradiation
of high interferential illumination.35 The formed superstructures
show a large amplitude, which is even larger than the initial film
thickness. In a previous paper, we have reported that a photo-
induced self-structured surface pattern can be induced on films of
a star-shaped molecular azo glass.36 Because of the well-defined
structures, molecular azo glasses can be a desirable type of material
for studying the influence of the molecular structures and light
irradiation conditions on the self-structured surface pattern for-
mation. However, to our knowledge, a report concerning a
systematic study in this area is still lacking in the literature.
In this work, three series of star-shaped molecular azo glasses
with different molecular structures were synthesized. Using the
amorphous molecular materials, the self-structured surface pat-
tern formation was studied by irradiating films of the materials
with a laser beam at three different wavelengths (488, 532, and
589 nm). The influences of the molecular structures and light
irradiation conditions on the surface pattern formation behavior
were investigated in detail.
carried out using TA Instruments DSC 2920 and TGA 2050 systems
with a heating rate of 10 °C/min in a nitrogen atmosphere. UVꢀvis
absorption spectra were recorded on a Perkin-Elmer Lamba Bio-40
spectrophotometer.
N-Ethyl-3,5-dimethylaniline. 3,5-Dimethylaniline (12.1 g, 0.1 mol)
and bromoethane (10.9 g, 0.1 mol) were dissolved in THF to form a
homogeneous solution. Anhydrous potassium carbonate (15.2 g, 0.11 mol)
and a small amount of potassium iodide were added to the solution. The
reaction was carried out at 40 °C for 6 h under the protection of nitrogen
gas. After the reaction, the mixture was filtrated and the solvent in the filtrate
was removed by rotary evaporation. The residue was purified by column
chromatography using a mixture of ethyl acetate and petroleum ether (v/v =
1/12) as the eluent. The final product was obtained as a yellow liquid. Yield:
49%. MS: m/z [M]+ 149.19, calcd 149.12. 1H NMR (DMSO-d6): δ (ppm)
1.13 (m, 3H), 2.12 (s, 6H), 2.97 (m, 2H), 5.24 (br, 1H), 6.15 (s, 3H).
2-((3,5-Dimethylphenyl)amino)ethanol. This compound was
synthesized by the reaction between 3,5-dimethylaniline (12.1 g, 0.1 mol)
and 2-chloroethanol (8.06 g, 0.1 mol) via a procedure similar to that for
preparing N-ethyl-3,5-dimethylaniline with a reaction temperature of
90 °C and a reaction time of 8 h. Yield: 53%. MS: m/z [M]+ 164.96, calcd
165.12. 1H NMR (DMSO-d6): δ (ppm) 2.13 (s, 6H), 3.06 (m, 2H), 3.54
(m, 2H), 4.63 (m, 1H), 5.18 (br, 1H), 6.17 (s, 1H), 6.19 (s, 2H).
Core Precursors Tr-AN, Tr-35AN, and Tr-H35AN. The core
precursor Tr-AN was synthesized by a method reported previously.36
In the reaction, 1,3,5-triglycidyl isocyanurate (2.97 g, 0.01 mol) and
N-methylaniline (4.28 g, 0.04 mol) were mixed and slowly heated in the
reactor. After the solid was melted, the mixture was continuously stirred
and kept at 110 °C for 8 h. The product was dissolved in THF (50 mL)
and then precipitated with petroleum ether. The solid was collected by
filtration and dried in a vacuum oven at 70 °C for 24 h. The crude
product was further purified by column chromatography (CH2Cl2, the
first component). The core precursors Tr-35AN and Tr-H35AN were
synthesized by reaction of 1,3,5-triglycidyl isocyanurate with N-ethyl-
3,5-dimethylaniline and 2-((3,5-dimethylphenyl)amino)ethanol via a
procedure similar to that for preparing Tr-AN. The analytical result for
Tr-AN has been reported before,36 and analytical results for Tr-35AN
and Tr-H35AN are given below.
Data for Tr-35AN. Yield: 83%. MS: m/z [M]+ 744.52, calcd 744.46.
1H NMR (DMSO-d6): δ (ppm) 1.03 (m, 3H), 2.15 (s, 6H), 3.15 (m,
1H), 3.30 (m, 2H), 3.38 (m, 1H), 3.71 (m, 1H), 3.90 (m, 1H), 4.04 (m,
1H), 4.89 (d, 1H), 6.19 (s, 1H), 6.25 (s, 2H).
Data for Tr-H35AN. Yield: 85%. MS: m/z [M]+ 792.34, calcd 792.44.
1H NMR (DMSO-d6): δ (ppm) 2.14 (s, 6H), 3.20 (m, 1H), 3.33 (m,
1H), 3.42ꢀ3.53 (m, 5H), 3.71 (m, 1H), 3.90 (m, 1H), 4.08 (m, 1H),
4.67 (m, 1H), 4.99 (d, 1H), 6.20 (s, 1H), 6.26 (s, 2H).
Azo Compounds Tr-AZ-Cl, Tr-AZ-CN, Tr-AZ-NT, Tr-35AZ-
Cl, Tr-35AZ-CN, Tr-35AZ-NT, Tr-H35AZ-Cl, Tr-H35AZ-CN, and
Tr-H35AZ-NT. The azo compounds were synthesized by azo-coupling
reactions.36 The synthesis of Tr-AZ-Cl is given here as a typical example.
4-Chloroaniline (0.64 g, 5 mmol) was mixed with sulfuric acid (0.5 mL)
and glacial acetic acid (12 mL). Diazonium salt was prepared by slowly
adding an aqueous solution of sodium nitrite (0.4 g, 5.78 mmol, in 1 mL
of water) to the 4-chloroaniline solution. The mixture was stirred at 5 °C
for 5 min and then added dropwise to a solution of Tr-AN (0.618 g,
1 mmol) in N,N-dimethylformamide (DMF; 60 mL). The solution was
stirred at 0 °C for 8 h and then poured into plenty of water. The
precipitate collected by filtration was dissolved in THF, and it was again
precipitated with petroleum ether. The final product was vacuum-dried
at 70 °C for 24 h. The azo compounds Tr-AZ-CN and Tr-AZ-NT were
synthesized by the azo-coupling reactions between Tr-AN and diazo-
nium salts of 4-aminobenzonitrile and 4-nitroaniline via a procedure
similar to that for preparing Tr-AZ-Cl. The azo compounds Tr-35AZ-Cl,
Tr-35AZ-CN, Tr-35AZ-NT, Tr-H35AZ-Cl, Tr-H35AZ-CN, and Tr-
H35AZ-NT were synthesized by the azo-coupling reactions of Tr-35AN
’ EXPERIMENTAL SECTION
Materials. 3,5-Dimethylaniline, bromoethane, 2-chloroethanol, 4-chloro
aniline, 4-aminobenzonitrile, 4-nitroaniline, and 1,3,5-triglycidyl isocyanurate
were purchased from Alfa Aesar. All the other materials, reagents, and solvents
were commercially available products, which were used as received without
further purification.
Characterization. Infrared spectra were determined using a Nico-
let 560-IR FT-IR spectrophotometer by incorporating samples in KBr
1
disks. H NMR spectra were recorded using a JEOL JNM-ECA600
NMR spectrometer. Mass spectra were determined by an Agilent 6300
ion trap LC/MS system. Thermal analyses of the compounds were
12667
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
and Tr-H35AN with the diazonium salts of 4-chloroaniline, 4-amino-
benzonitrile, and 4-nitroaniline, respectively, via a procedure similar to
that for preparing Tr-AZ-Cl. Analytical results for Tr-AZ-CN and Tr-
AZ-NT have been given before.36 Analytical results for the other newly
synthesized azo compounds are given below.
proper time period (typically 30 min). After the light irradiation, the
surfaces of the films were examined by using an atomic force microscope
(Nanoscope IIIa) in the tapping mode.
’ RESULTS AND DISCUSSION
Data for Tr-AZ-Cl. Yield: 87%. DSC: Tg = 98 °C. MS: m/z [M]+
1032.59, calcd 1032.31. 1H NMR (DMSO-d6): δ (ppm) 3.11 (s, 3H),
3.34 (m, 1H), 3.69 (m, 1H), 3.78 (m, 1H), 3.96 (m, 1H), 4.16 (m, 1H),
5.17 (d, 1H), 6.84 (d, 2H), 7.55 (d, 2H), 7.75 (m, 4H). IR (KBr, cmꢀ1):
3350, 2954, 2908, 1693, 1601, 1514, 1460, 1417, 1375, 1313, 1242, 1209,
1142, 1086, 1009, 949, 835, 760, 544.
Nine star-shaped azo compounds (Tr-AZ-Cl, Tr-AZ-CN, Tr-
AZ-NT, Tr-35AZ-Cl, Tr-35AZ-CN, Tr-35AZ-NT, Tr-H35AZ-
Cl, Tr-H35AZ-CN, and Tr-H35AZ-NT) were used to study the
photoinduced self-structured surface pattern formation. The
synthetic route and chemical structure of all azo compounds
are given in Scheme 1. Among them, Tr-AZ-CN and Tr-AZ-NT
have been synthesized and reported in our previous paper.36 For
these two azo compounds, the formation of the self-structured
surface patterns was only observed on the Tr-AZ-CN film upon
irradiation with the laser beam at 532 nm.36 In the current study,
the newly synthesized molecular azo glasses together with Tr-
AZ-CN and Tr-AZ-NT were used to study the structureꢀ
property relationship of the photoinduced surface pattern for-
mation. The correlation with the excitation wavelength was
studied upon irradiation with the laser beams at three different
wavelengths.
Data for Tr-35AZ-Cl. Yield: 88%. DSC: Tg = 92 °C. MS: m/z [M]+
1158.51, calcd 1158.45. 1H NMR (DMSO-d6): δ (ppm) 1.11 (m, 3H),
2.47 (s, 6H), 3.30 (m, 1H), 3.50ꢀ3.55 (m, 3H), 3.80 (m, 1H), 3.99 (m,
1H), 4.15 (m, 1H), 5.10 (d, 1H), 6.47 (s, 2H), 7.53 (d, 2H), 7.66 (d,
2H). IR (KBr, cmꢀ1): 3444, 2966, 2920, 1695, 1599, 1495, 1462, 1416,
1371, 1344, 1286, 1124, 1088, 835, 766, 521.
Data for Tr-35AZ-CN. Yield: 87%. DSC: Tg = 123 °C. MS: m/z [M]+
1131.63, calcd 1131.56. 1H NMR (DMSO-d6): δ (ppm) 1.12 (m, 3H),
2.50 (s, 6H), 3.31 (m, 1H), 3.50ꢀ3.57 (m, 3H), 3.80 (m, 1H), 4.00 (m,
1H), 4.15 (m, 1H), 5.13 (d, 1H), 6.50 (s, 2H), 7.76 (d, 2H), 7.90 (d,
2H). IR (KBr, cmꢀ1): 3481, 2968, 2922, 2224, 1693, 1597, 1498, 1462,
1367, 1344, 1309, 1286, 1273, 1176, 1142, 1122, 1076, 987, 845,
766, 553.
Synthesis and Characterization. The synthetic route of the
azo compounds includes the preparation of star-shaped core
precursors and the azo-coupling reactions of the precursors
(Scheme 1). N-Ethyl-3,5-dimethylaniline and 2-((3,5-dimethyl-
phenyl)amino)ethanol were synthesized through the nucleophi-
lic substitution of 3,5-dimethylaniline with bromoethane and
2-chloroethanol, respectively. Three star-shaped core precursors
(Tr-AN, Tr-35AN, Tr-H35AN) were synthesized through the
ring-opening reactions of 1,3,5-triglycidyl isocyanurate (TGIC)
with the aniline derivatives N-methylaniline, N-ethyl-3,5-di-
methylaniline, and 2-((3,5-dimethylphenyl)amino)ethanol, re-
spectively. Each series of azo compounds was obtained by intro-
ducing three types of azo chromophores through the azo-
coupling reactions of the core precursor with the corresponding
diazonium salts in DMF. The chemical structure and abbreviated
names of the azo compounds are shown in Scheme 1. The first
two parts of the names specify the azo compounds obtained from
the core precursors Tr-AN, Tr-35AN, and Tr-H35AN, and the
last part makes the distinction of the electron-withdrawing
groups on the azo chromophores. The series of azo compounds
based on Tr-35AN contain 3,5-dimethyl substituents on the
azobenzene moieties, which is different from those obtained
from Tr-AN. The series of azo compounds obtained from Tr-
H35AN bear an extra hydroxyl on the ethyl group compared with
the azo compounds obtained from Tr-35AN. The structures of
the azo compounds were confirmed by 1H NMR, FT-IR,
UVꢀvis, and mass spectrometry.
Data for Tr-35AZ-NT. Yield: 84%. DSC: Tg = 122 °C. MS: m/z [M]+
1191.61, calcd 1191.53. 1H NMR (DMSO-d6): δ (ppm) 1.11 (m, 3H),
2.51 (s, 6H), 3.34 (m, 1H), 3.54ꢀ3.62 (m, 3H), 3.82 (m, 1H), 3.99 (m,
1H), 4.16 (m, 1H), 5.16 (d, 1H), 6.52 (s, 2H), 7.78 (d, 2H), 8.28 (d,
2H). IR (KBr, cmꢀ1): 3521, 2968, 2920, 1695, 1601, 1514, 1462, 1335,
1271, 1174, 1144, 1124, 1105, 987, 858, 766.
Data for Tr-H35AZ-Cl. Yield: 85%. DSC: Tg = 103 °C. MS: m/z [M]+
1206.45, calcd 1206.44. 1H NMR (DMSO-d6): δ (ppm) 2.47 (s, 6H),
3.34 (m, 1H), 3.51ꢀ3.63 (m, 5H), 3.79 (m, 1H), 3.97 (m, 1H), 4.18 (m,
1H), 4.79 (m, 1H), 5.16 (d, 1H), 6.49 (s, 2H), 7.51 (d, 2H), 7.66 (d,
2H). IR (KBr, cmꢀ1): 3284, 2956, 2918, 1693, 1597, 1462, 1348, 1273,
1163, 1088, 1034, 1007, 985, 835, 766, 521.
Data for Tr-H35AZ-CN. Yield: 86%. DSC: Tg = 125 °C. MS: m/z
[M]+ 1179.55, calcd 1179.54. 1H NMR (DMSO-d6): δ (ppm) 2.50 (s,
6H), 3.38 (m, 1H), 3.54ꢀ3.63 (m, 5H), 3.67 (m, 1H), 3.80 (m, 1H),
3.99 (m, 1H), 4.19 (m, 1H), 4.82 (m, 1H), 5.18 (d, 1H), 6.53 (s, 2H),
7.77 (d, 2H), 7.90 (d, 2H). IR (KBr, cmꢀ1): 3377, 2956, 2918, 2225,
1693, 1597, 1464, 1350, 1271, 1171, 1142, 1099, 1045, 982, 843,
766, 553.
Data for Tr-H35AZ-NT. Yield: 82%. DSC: Tg = 133 °C. MS: m/z [M]+
1239.57, calcd 1239.51. 1H NMR (DMSO-d6): δ (ppm) 2.52 (s, 6H),
3.40 (m, 1H), 3.57ꢀ3.63 (m, 5H), 3.71 (m, 1H), 3.80 (m, 1H), 3.99 (m,
1H), 4.20 (m, 1H), 4.83 (m, 1H), 5.20 (d, 1H), 6.55 (s, 2H), 7.80 (d,
2H), 8.28 (d, 2H). IR (KBr, cmꢀ1): 3311, 2956, 2918, 1693, 1601, 1514,
1464, 1335, 1269, 1167, 1144, 1103, 982, 856, 756, 690.
Film Preparation. Solid films of the azo compounds with smooth
surfaces were prepared by spin-coating. The homogeneous solutions of
the azo compounds were obtained by dissolving a suitable amount of the
azo compounds in DMF. After being filtered through 0.45 μm mem-
branes, the solutions were spin-coated onto clean glass slides. The film
thicknesses were controlled to be in a range from 400 to 600 nm by
adjusting the solution concentrations and the spinning speed. The films
obtained from the spin-coating were dried at 60 °C under vacuum for 48 h
before use.
All three series of the star-shaped azo compounds obtained
from Tr-AN, Tr-35AN, and Tr-H35AN show typical behavior of
the amorphous materials. The glass transition temperatures (Tg)
and decomposition temperatures (Td) of the star-shaped azo
compounds obtained from thermal analyses are given in Table 1.
The Tg values of the azo compounds are significantly higher than
those of the corresponding core precursors because of the
existence of the azo chromophores. On the other hand, com-
pared with the corresponding core precursors, the Td values of
the azo compounds are reduced due to the lower thermal stability
of the azo bonds. The azo compounds were easily dissolved in
polar organic solvents such as tetrahydrofuran (THF), DMF, and
N,N-dimethylacetamide (DMAc). Solid thin films with smooth
Laser Light Irradiation. A linearly or circularly polarized beam
from an Ar+ laser (488 nm) and diode-pumped frequency-doubled solid-
state lasers (532 and 589 nm) was used as the excitation light. The laser
beam with proper intensity was obtained after being spatially filtered,
expanded, and collimated. The laser beam (typically with an intensity of
200 mW/cm2) was perpendicularly incident to the film surfaces for a
12668
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Scheme 1. Synthetic Route of the Star-Shaped Azo Compounds
surfaces could be obtained through spin-coating by using the azo
compound solutions.
withdrawing group. This section gives the results of three star-
shaped azo compounds with chloro as the electron-withdrawing
group on the azo chromophores. The results were obtained by
irradiating the solid films with a uniform normal-incident laser
beam (200 mW/cm2) at three different wavelengths (488, 532,
and 589 nm).
Figure 1 gives UVꢀvis absorption spectra of the spin-coated
films of the three series of azo compounds. The spectra exhibit
typical characteristics of the pseudo-stilbene-type azo chromo-
phores, where the absorption bands corresponding to the πꢀπ*
transition appear in the visible spectrum region. The λmax values
of the πꢀπ* transition for the azo compounds (both spin-coated
films and DMF solutions) are listed in Table 1. The λmax values of
the azo compounds are strongly affected by the electron-with-
drawing groups on the azo chromophores. As the electron-
withdrawing ability increases, the λmax values increase in the
order of chloro, cyano, and nitro substituents. The λmax values of
the azo compounds in DMF solutions show significant red shifts
compared with those of spin-coated films, which can be attrib-
uted to the strong solvatochromic effect of DMF.
Molecular Azo Glasses with a Chloro Electron-Withdraw-
ing Group. When irradiated at different wavelengths, the
molecular azo glasses with different λmax values show distinct
pattern formation behavior. As λmax values are mainly deter-
mined by the electron-withdrawing groups on the azo chromo-
phores, the behavior of the photoinduced self-structured pattern
formation will be discussed separately for each type of electron-
Figure 2 shows typical AFM images of the photoinduced
surface patterns formed on the Tr-AZ-Cl film upon laser irra-
diation at 488 nmfor 30 min. After the irradiation, the film surfaces
show regularly spaced pillars with some localized hexagonal
arrangement. Obvious differences can be seen for the surface
patterns formed after irradiations of linearly and circularly polarized
laser beams with the same light intensity and irradiation time.
Upon irradiation with linearly polarized light, the alignment
of the localized hexagonal pattern shows correlation with the
polarization direction (Figure 2a,b), which is confirmed by the
two-dimensional Fourier transform (2D-FT) of the AFM image
(Figure 2c). The period of the regularly spaced pillars is 450 (
15 nm, which is obtained along (60° with respect to the light
polarization direction. The pattern formed by the circularly
polarized light irradiation does not show the alignment correla-
tion (Figure 2d,e). The 2D-FT analysis of the AFM image
verifies the 2D isotropic alignment of the pillars (Figure 2f).
12669
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Table 1. Thermal and Spectral Parameters of the Star-Shaped
Azo Compounds
Tg
Td
λmax (nm)
λmax (nm)
compound
Tr-AN
(°C)
(°C)
(DMF solution)
(spin-coated film)
38
98
310
268
277
250
286
255
247
237
313
263
240
233
Tr-AZ-Cl
430
463
494
409
424
454
Tr-AZ-CN
120
116
43
Tr-AZ-NT
Tr-35AN
Tr-35AZ-Cl
Tr-35AZ-CN
Tr-35AZ-NT
Tr-H35AN
Tr-H35AZ-Cl
Tr-H35AZ-CN
Tr-H35AZ-NT
92
424
451
490
404
433
463
123
122
55
103
125
133
423
450
487
399
425
453
The saturated height and period of the surface patterns for Tr-
AZ-Cl and other molecular azo glasses with chloro electron-
withdrawing groups are summarized in Table 2. For Tr-AZ-Cl,
the saturated amplitude of the surface patterns is 59 ( 11 nm for
irradiation with linearly polarized light. Meanwhile, the saturated
amplitude of the surface patterns induced by circularly polarized
light is 85 ( 15 nm. The above observations of the influence of
light polarization on self-structured surface pattern formation are
consistent with those reported for DR1MA/MMA 35/65 and
molecular azo glass.19,36
With irradiation with a linearly polarized laser beam at 532 nm
under similar conditions, a self-structured surface pattern was
also observed to form on the Tr-AZ-Cl film. The period of the
regularly spaced pillars formed on the Tr-AZ-Cl film is 490 (
20 nm along the directions of (60° with respect to the
polarization direction. The saturated amplitude of the surface
patterns is 35 ( 8 nm, which is smaller than that obtained by
irradiation with the 488 nm laser beam. When the film of Tr-AZ-
Cl was irradiated with the linearly polarized laser beam (200
mW/cm2) at 589 nm for 30 min, no self-structured surface
pattern formation could be observed.
Figure 3 shows the patterns formed on the films of Tr-35AZ-
Cl after irradiation with the linearly polarized laser beam (200
mW/cm2) at 488 nm for 30 min. Similar to Tr-AZ-Cl, surface
pattern formation can be observed on the film of Tr-35AZ-Cl
(Figure 3a,b). The saturated amplitude of the pillarlike structures
for the Tr-35AZ-Cl film is only 12 ( 3 nm, which is much smaller
than that of Tr-AZ-Cl when irradiated with the 488 nm light. The
smaller saturated amplitude can be attributed to the steric
hindrance of the bulky 3,5-dimethyl groups and the inhibition
of the transꢀcisꢀtrans isomerization of the adjacent azo bond.
The space period of the pillars is 455 ( 15 nm. For comparison,
the film of Tr-35AZ-Cl was also irradiated with the linearly
polarized laser beam (200 mW/cm2) at 532 nm for 30 min. The
self-structured surface pattern forms on the film of Tr-35AZ-Cl
with a saturated amplitude of 42 ( 9 nm, which is even slightly
higher than that of Tr-AZ-Cl when irradiated with 532 nm light.
The possible reason why there is no obvious steric hindrance
after irradiation with 532 nm light is given in the Discussion
by considering the different photoisomerization pathways of the
azo chromophores. The space period of the pillarlike structures is
490 ( 20 nm. When the film of Tr-35AZ-Cl was irradiated with the
Figure 1. UVꢀvis spectra of the molecular azo compounds as spin-
coated films: (a) azo compounds obtained from Tr-AN; (b) azo
compounds obtained from Tr-35AN; (c) azo compounds obtained
from Tr-H35AN.
linearly polarized laser beam (200 mW/cm2) at 589 nm for 30 min,
no self-structured surface pattern formation could be observed.
For the Tr-H35AZ-Cl film, no obvious self-structured surface
pattern formation was observed after irradiation with 488 nm
light under the same conditions. Only shallow irregular fluctua-
tions with an amplitude of less than 6 nm appeared on the surface
(Figure 3c,d). Even after irradiation with the laser beam with a
much higher intensity and for a longer irradiation time (400 mW/cm2,
60 min), no regular self-structured surface patterns could be
seen on the Tr-H35AZ-Cl film. The main difference between
Tr-35AZ-Cl and Tr-H35AZ-Cl is the extra hydroxyl on the ethyl
group. This observation indicates that the extra hydroxyl groups
in the Tr-H35AZ-Cl structure can further inhibit the surface
pattern formation. This can be attributed to the strong inter-
molecular interaction through hydrogen bonding of the hydro-
xyl groups. Similarly, when the films of Tr-H35AZ-Cl were
irradiated with the linearly polarized laser beam (200 mW/cm2)
at 532 nm for 30 min, no self-structured surface pattern
formation could be observed on the film. Like Tr-AZ-Cl and
Tr-35AZ-Cl, after irradiation at 589 nm with the linearly polar-
ized laser beam (200 mW/cm2) for 30 min, no self-structured
surface pattern formation could be observed for the Tr-H35AZ-Cl
film. In these cases, the irradiation wavelength is much longer
than the λmax values of the πꢀπ* transition bands and even
away from the absorption band tails at the longer wavelength
side (Figure 1).
12670
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Figure 2. Typical AFM images (10 μm ꢁ 10 μm) and 2D-FT images of the photoinduced surface patterns formed on Tr-AZ-Cl films upon irradiation
with a normal-incident laser beam at 488 nm and different polarizations: (a) linearly polarized, 2D view; (b) linearly polarized, 3D view; (c) linearly
polarized, 2D-FT image; (d) circularly polarized, 2D view; (e) circularly polarized, 3D view; (f) circularly polarized, 2D-FT image. The light intensity
was 200 mW/cm2, and the irradiation time was 30 min.
Table 2. Saturated Amplitude and Space Period of the Surface Patterns on the Molecular Azo Glasses with Chloro as the Electron-
Withdrawing Group Obtained under Different Light Irradiation Conditions
azo compound
wavelength (nm)
intensity (mW/cm2)
time (min)
state of polarization
amplitude (nm)
period (nm)
Tr-AZ-Cl
488
488
532
488
532
488
488
532
200
200
200
200
200
200
400
200
30
30
30
30
30
30
60
30
linearly
circularly
linearly
linearly
linearly
linearly
linearly
linearly
59 ( 11
85 ( 15
35 ( 8
12 ( 3
42 ( 9
450 ( 15
Tr-AZ-Cl
Tr-AZ-Cl
490 ( 20
455 ( 15
490 ( 20
Tr-35AZ-Cl
Tr-35AZ-Cl
Tr-H35AZ-Cl
Tr-H35AZ-Cl
Tr-H35AZ-Cl
Molecular Azo Glasses with a Cyano Electron-Withdraw-
ing Group. For the azo compounds with cyano as the electron-
withdrawing group, the photoinduced self-structured surface
pattern formation was studied by the same method. The
saturated heights and periods of the surface patterns for the
molecular azo glasses are summarized in Table 3. When the films
of Tr-AZ-CN, Tr-35AZ-CN, and Tr-H35AZ-CN were irradiated
with the linearly polarized laser beam (200 mW/cm2) at 488 nm
for 30 min, the surface pattern could only be observed to form on
the film of Tr-AZ-CN with a saturated amplitude of the surface
patterns of 20 ( 4 nm and a space period of 420 ( 15 nm.
With irradiation with the laser beam (200 mW/cm2) at
532 nm, the self-structured surface patterns can be efficiently
induced on films of Tr-AZ-CN as reported by us before.36
Compared with Tr-AZ-CN, the surface pattern formation on
the film of Tr-35AZ-CN is less efficient upon irradiation with
532 nm light under the same conditions. Figure 4 shows AFM
images of the photoinduced surface patterns formed on the Tr-
35AZ-CN film upon laser irradiation (200 mW/cm2) at 532 nm
for 30 min. The irradiations with the linearly and circularly
polarized laser beams also show different influences on the pattern
formation. With irradiation with the linearly polarized laser, the
alignment of the localized hexagonal pattern shows correlation
with the polarization direction (Figure 4aꢀc). The saturated
amplitude of the surface pattern is 25 ( 4 nm with a space period
of 470 ( 20 nm. Upon irradiation with the circularly polarized
laser, the pattern does not show the alignment correlation
(Figure 4dꢀf). The saturated amplitude of the surface patterns
is 85 ( 18 nm, which is significantly higher than the case for
linearly polarized light irradiation. For Tr-35AZ-CN, the satu-
rated amplitude of the surface patterns is obviously smaller than
that of Tr-AZ-CN, especially for the film irradiated with the
linearly polarized laser.
When the film of Tr-H35AZ-CN was irradiated with the
linearly polarized laser beam at 532 nm (200 mW/cm2, 30 min),
no self-structured surface pattern formation could be observed.
However, when the Tr-H35AZ-CN film was irradiated at 532 nm
with a much higher intensity (400 mW/cm2, 30 min), self-
structured surface pattern formation was observed. The pillarlike
structures show a saturated amplitude of 50 ( 10 nm with a space
12671
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Figure 3. AFM images (10 μm ꢁ 10 μm) of the photoinduced surface patterns formed on Tr-35AZ-Cl and Tr-H35AZ-Cl films upon irradiation
at 488 nm: (a) Tr-35AZ-Cl, 2D view; (b) Tr-35AZ-Cl, 3D view; (c) Tr-H35AZ-Cl, 2D view; (d) Tr-H35AZ-Cl, 3D view. The light intensity was
200 mW/cm2, and the irradiation time was 30 min.
Table 3. Saturated Amplitude and Space Period of the Surface Patterns on the Molecular Azo Glasses with Cyano as the Electron-
Withdrawing Group Obtained under Different Light Irradiation Conditions
azo compound
wavelength (nm)
intensity (mW/cm2)
time (min)
state of polarization
amplitude (nm)
period (nm)
Tr-AZ-CN
488
532
532
488
532
532
488
532
532
200
200
200
200
200
200
200
200
400
30
15
15
30
30
30
30
30
30
linearly
linearly
circularly
linearly
linearly
circularly
linearly
linearly
linearly
20 ( 4
80
420 ( 15
440 ( 20
Tr-AZ-CN36
Tr-AZ-CN36
Tr-35AZ-CN
Tr-35AZ-CN
Tr-35AZ-CN
Tr-H35AZ-CN
Tr-H35AZ-CN
Tr-H35AZ-CN
120
25 ( 4
470 ( 20
85 ( 18
50 ( 10
490 ( 20
period of 490 ( 20 nm. When films of Tr-AZ-CN, Tr-35AZ-CN,
and Tr-H35AZ-CN were irradiated at 589 nm with the linearly
polarized laserbeam (200 mW/cm2) for30min, noself-structured
surface pattern formation could be observed, which is similar to
the cases for Tr-AZ-Cl, Tr-35AZ-Cl, and Tr-H35AZ-Cl. This can
also be attributed to the very weak absorption at the irradiation
wavelength as shown in Figure 1.
Molecular Azo Glasses with a Nitro Electron-Withdrawing
Group. Tr-AZ-NT, Tr-35AZ-NT, and Tr-H35AZ-NT bear
strong electron-withdrawing nitro groups on the azo chromo-
phores. Due to the strong electron-withdrawing effect, the mol-
ecular azo glasses possess significantly larger λmax values for the
πꢀπ* transition. Therefore, the molecular azo glasses show strong
absorption for light with wavelengths of 488 and 532 nm and have
some absorption even for the 589 nm light. However, different
from the above-mentioned azo compounds, these three molecular
azo glasses did not show the ability to form a surface pattern when
irradiated with the linearly polarized laser beam (200 mW/cm2)
at 488, 532, and 589 nm for 30 min. The possible reason for this
observation is given in the Discussion.
Dynamic Process of Pattern Formation. As discussed above,
Tr-AZ-Cl exhibits the typical self-structured surface pattern
formation behavior when irradiated with 488 nm light. There-
fore, the molecular azo glass was used to study the dynamic
process of pattern formation. Figures 5 and 6 reveal the gradual
pattern formation processes on Tr-AZ-Cl films when irradiated
with linearly and circularly polarized laser beams at 488 nm for
different time periods. By closely examining the AFM images
given in Figure 5, it can be seen that, after irradiation with the
linearly polarized laser for 30 min, the pillarlike structures
obviously merge in two directions about (60° to the light
polarization direction. On the contrary, no such deformation is
observed after irradiation with the circularly polarized laser for
the same time. The pattern formation seems to be in a more
12672
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Figure 4. Typical AFM images (10 μm ꢁ 10 μm) and 2D-FT images of the photoinduced surface patterns formed on Tr-35AZ-CN films upon
irradiation with a normal-incident laser beam at 532 nm and different polarizations: (a) linearly polarized, 2D view; (b) linearly polarized, 3D view;
(c) linearly polarized, 2D-FT image; (d) circularly polarized, 2D view; (e) circularly polarized, 3D view; (f) circularly polarized, 2D-FT image. The light
intensity was 200 mW/cm2, and the irradiation time was 30 min.
Figure 5. AFM images (10 μm ꢁ 10 μm) of the photoinduced surface patterns on Tr-AZ-Cl films upon irradiation for different times. The incident laser
beam (488 nm, 200 mW/cm2) was linearly polarized. The irradiation time was (a) 5 min, (b) 10 min, (c) 15 min, (d) 20 min, (e) 30 min, and (f) 40 min.
complicated and random manner. This observation further
verifies that the light polarization condition can cause the obvious
difference in the pattern formation and surface deformation
manner. Figure 7 gives the corresponding relationship between
the modulation amplitude and irradiation time. It can be seen
that, after a slow initial stage, the self-structured pattern forma-
tion exhibits a rapid increase in the first 15 min. After that, the
pattern growth rate is obviously slowed. With irradiation with the
circularly polarized laser beam, the self-structured pattern for-
mation reaches the saturated level in about 20 min. With
irradiation with the linearly polarized laser beam, the amplitude
still gradually increases after 20 min and becomes saturated in
30ꢀ40 min.
Discussion. The above results indicate that self-structured sur-
face pattern formation is a process closely related to the molecular
structure and excitation wavelength. The structural factors studied
here include the type of electron-withdrawing groups, the 3,
5-dimethyl substitution, and the extra hydroxyl group. The influence
of the light wavelength is also correlated with some of the structural
factors. All these factors and correlation are discussed below.
The electron-withdrawing groups on the azo chromophores
show a significant influence on the surface pattern formation. Tr-
AZ-Cl, Tr-35AZ-Cl, Tr-AZ-CN, and Tr-35AZ-CN, which con-
tain chloro and cyano as the electron-withdrawing groups, show
the ability to form self-structured surface patterns when irra-
diated with light at the proper wavelengths (488 and 532 nm).
12673
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
Figure 6. AFM images (10 μm ꢁ 10 μm) of the photoinduced surface patterns on Tr-AZ-Cl films upon irradiation for different times. The incident
laser beam (488 nm, 200 mW/cm2) was circularly polarized. The irradiation time was (a) 5 min, (b) 10 min, (c) 15 min, (d) 20 min, (e) 30 min, and
(f) 40 min.
between the λmax of the trans πꢀπ* transition and the λmax of the
cis nꢀπ* transition, the rate of SRG formation is increased at a
longer laser excitation wavelength. For the surface pattern
formation studied here, the correlation with the excitation
wavelength could also be attributed to the different transꢀcis
isomerization efficiencies connected to the excitation at the
absorption band positions. Tr-35AZ-Cl might not be a true
exception to this rule, although it does form the surface pattern
more efficiently when irradiated with 532 nm light than with
488 nm light. The possible reason for this phenomenon is given
below by considering the substitution effect.
Figure 7. Amplitude of the surface patterns formed on a Tr-AZ-Cl film
The 3,5-dimethyl substitution on the azo chromophores shows
as a function of the irradiation time upon irradiation with linearly and
more complicated influences on self-structured surface pattern for-
circularly polarized laser beams. The wavelength of the laser beams was
488 nm, and the light intensity was 200 mW/cm2.
mation. In most cases, the 3,5-dimethyl substitution on the azo
chromophores obstructs the self-structured surface pattern forma-
tion, which results in a lower saturated amplitude. This can be seen
for Tr-35AZ-Cl irradiated with 488 nm light and Tr-35AZ-CN
irradiated with 488 and 532 nm light. This effect can be attributed to
the steric hindrance of the dimethyl substitution to the transꢀcis
isomerization of azo chromophores through the rotation mechan-
ism (the twist around the C—NdN—C dihedral angle).3,40 As
mentioned above, upon irradiation with 532 nm light, the amplitude
of the pillarlike structures on the Tr-35AZ-Cl film is similar to that of
Tr-AZ-Cl. The 3,5-dimethyl substitution does not show obvious
inhibition of pattern formation. A possible explanation is that
irradiation at this wavelength could mainly cause nꢀπ* transition
and the transꢀcis isomerization might go through the inversion
pathway (the inversion of one or both of the CNN angles through a
linear transition state).3,40 In this case, the dimethyl substitution
could not show the steric hindrance to the photoinduced transꢀcis
isomerization of the azo chromophores. Considering the model
based on competition between excitation and relaxation,39 if the
λmax values of both trans nꢀπ* transition and cis nꢀπ* transition
are closer to 532 nm, irradiation at this wavelength can cause
efficient transꢀcis isomerization. For this reason, Tr-35AZ-Cl is
more efficient at forming the surface pattern when irradiated with
532 nm light compared with 488 nm light.
For the molecular azo glasses bearing cyano as the electron-
withdrawing group, the λmax values of the πꢀπ* transition are
red-shifted about 30 nm compared with those of the molecular
azo glasses bearing a chloro group. Therefore, the molecular azo
glasses show distinct efficiencies in response to the different
excitation wavelengths. When the irradiation time is the same,
the amplitude of the pillarlike structures can be used to characterize
the relative efficiency of pattern formation. For Tr-AZ-Cl,
self-structured surface pattern formation is more efficiently
induced by irradiation with 488 nm light. On the other hand,
pattern formation on the Tr-AZ-CN and Tr-35AZ-CN films is
more efficiently induced by irradiation with 532 nm light.
Comparing this observation with the UVꢀvis spectra of the
compounds (Figure 1), it can be inferred that when the excitation
wavelength is between the λmax of the πꢀπ* transition and the
absorption band tail at the longer wavelength side, light irradia-
tion can cause surface pattern formation more efficiently. A
similar influence of the excitation wavelength on the SRG
formation efficiency has been reported before.37ꢀ39 It has been
indicated that the transꢀcis photoisomerization efficiency de-
pends on the competition between excitation (πꢀπ* or nꢀπ*
transition) and relaxation.39 The transꢀcis isomerization yield of
the nꢀπ* excitation is higher than that of the πꢀπ* excitation.
Therefore, under conditions of adequate excitation at a wavelength
Intermolecular interaction shows obvious obstruction to the
surface pattern formation. This can be seen for the molecular azo
glasses obtained from Tr-H35AN, which bear an extra hydroxyl
12674
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
on the ethyl group. After irradiation with the light (200 mW/cm2),
no pattern formation is observed for this series of azo com-
pounds, which can be attributed to the hydrogen bonds between
the molecules. This inference is partially supported by the obser-
vation that self-structured surface patterns could be induced on
the Tr-H35AZ-CN film when irradiated with a much higher
intensity light (400 mW/cm2). For molecular azo glasses con-
taining nitro substituents on the azo chromophore, no self-
structured surface patterns can be induced by irradiation at the
three wavelengths. The dipole moment of the azo chromophore
bearing nitro as the electron-withdrawing group is significantly
larger than that of the azo chromophore bearing cyano as the
electron-withdrawing group (9.1 D vs 6.4 D).39 It has been
reported that the strong dipolar interaction exists between the
4-amino-40-nitroazobenzene moieties of the side-chain azo
polymer.41 It is believed that the inhibition of the surface pattern
formation could be attributed to the strong dipoleꢀdipole inter-
action between the 4-amino-40-nitroazobenzene moieties. On
the other hand, it cannot be ruled out that, for Tr-AZ-NT, Tr-
35AZ-NT, and Tr-H35AZ-NT, the λmax values of the cis nꢀπ*
transitions are located at even longer wavelength positions. In
this case, the lower transꢀcis isomerization efficiency could also
result in the failure to form the self-structured surface pattern.
The structureꢀproperty relationship and correlation with the
wavelength discussed above can shed some new light on the
mechanism of self-structured surface pattern formation. First, the
above results suggest that the photoinduced transꢀcis isomer-
ization plays a critical role in the pattern formation process. One
point of direct evidence is that irradiation with 589 nm light
cannot cause pattern formation, which is away from the absorp-
tion bands and unable to cause transꢀcis isomerization. Second,
the driving force for pattern formation is quite weak, which can
be significantly obstructed by the noncovalent molecular inter-
action. Third, the difference with the LIPSS can also be seen from
the relationship between the excitation wavelength and the space
periods of the surface patterns. For the LIPSS, the ripple spacing
is approximately equal to the vacuum wavelength at normal
incidence.25,26 Although the space periods reported here show
correlation with the light wavelengths, they are always smaller
than the light wavelengths (Tables 2 and 3).
for pattern formation is quite weak, which can be obviously
restrained by the noncovalent molecular interaction.
’ ASSOCIATED CONTENT
S
Supporting Information. More characterization results of
b
the core precursors and star-shaped azo compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: wxg-dce@mail.tsinghua.edu.cn.
’ ACKNOWLEDGMENT
Financial support from the NSFC under Projects 91027024
and 51061130556 is gratefully acknowledged.
’ REFERENCES
(1) Viswanathan, N. K.; Kim, D. Y.; Bian, S. P.; Williams, J.; Liu, W.;
Li, L.; Samuelson, L.; Kumar, J.; Tripathy, S. K. J. Mater. Chem. 1999, 9
(9), 1941–1955.
(2) Delaire, J. A.; Nakatani, K. Chem. Rev. 2000, 100 (5), 1817–1845.
(3) Natansohn, A.; Rochon, P. Chem. Rev. 2002, 102 (11), 4139–4175.
(4) Cojocariu, C.; Rochon, P. Pure Appl. Chem. 2004, 76, 1479–1497.
(5) Rochon, P.; Batalla, E.; Natansohn, A. Appl. Phys. Lett. 1995, 66
(2), 136–138.
(6) Kim, D. Y.; Tripathy, S. K.; Li, L.; Kumar, J. Appl. Phys. Lett. 1995,
66 (10), 1166–1168.
(7) Che, P. C.; He, Y. N.; Wang, X. G. Macromolecules 2005, 38 (21),
8657–8663.
(8) Perschke, A.; Fuhrmann, T. Adv. Mater. 2002, 14 (11), 841–843.
(9) Kim, M. J.; Kumar, J.; Kim, D. Y. Adv. Mater. 2003, 15 (23),
2005–2008.
(10) Li, Y. B.; He, Y. N.; Tong, X. L.; Wang, X. G. J. Am. Chem. Soc.
2005, 127 (8), 2402–2403.
(11) Liu, B.; He, Y. N.; Wang, X. G. Langmuir 2006, 22 (24), 10233–
10237.
(12) Guo, M. C.; Xu, Z. D.; Wang, X. G. Langmuir 2008, 24 (6),
2740–2745.
(13) Goldenberg, L. M.; Gritsai, Y.; Kulikovska, O.; Stumpe, J. Opt.
Lett. 2008, 33 (12), 1309–1311.
(14) (a) Barrett, C. J.; Rochon, P.; Natansohn, A. J. Phys. Chem.
1996, 100, 8836–8842. (b) Barrett, C. J.; Rochon, P.; Natansohn, A.
J. Chem. Phys. 1998, 109, 1505–1516.
(15) (a) Kumar, J.; Li, L.; Jiang, X. L.; Kim, D. Y.; Lee, T. S.;
Tripathy, S. K. Appl. Phys. Lett. 1998, 72, 2096–2098. (b) Bian, S.;
Williams, J. M.; Kim, D. Y.; Li, L.; Balasubramanian, S.; Kumar, J.;
Tripathy, S. K. J. Appl. Phys. 1999, 86, 4498–4508.
(16) Lefin, P.; Fiorini, C.; Nunzi, J. M. Pure Appl. Opt. 1998, 7,
71–82.
(17) Pedersen, T. G.; Johansen, P. M.; Holme, N. C. R.; Ramanujam,
P. S.; Hilvsted, S. Phys. Rev. Lett. 1998, 80, 89–92.
(18) (a) Tanchak, O. M.; Barrett, C. J. Macromolecules 2005,
38, 10566–10570. (b) Yager, K. G.; Tanchak, O. M.; Godbout, C.;
Fritzsche, H.; Barrett, C. J. Macromolecules 2006, 39, 9311–9319.
(c) Yager, K. G.; Barrett, C. J. Macromolecules 2006, 39, 9320–9326.
(19) Hubert, C.; Fiorini-Debuisschert, C.; Maurin, I.; Nunzi, J. M.;
Raimond, P. Adv. Mater. 2002, 14, 729–732.
(20) Hubert, C.; Fiorini-Debuisschert, C.; Rocha, L.; Raimond, P.;
Nunzi, J. M. J. Opt. Soc. Am. B 2007, 24, 1839–1846.
(21) (a) Bolle, M.; Lazare, S.; Leblanc, M.; Wilmes, A. Appl. Phys.
Lett. 1992, 60 (6), 674–676. (b) Bolle, M.; Lazare, S. J. Appl. Phys. 1993,
73 (7), 3516–3524.
’ CONCLUSIONS
Three series of molecular azo glasses were prepared and used
for studying photoinduced surface pattern formation when
irradiated with laser beams at three different excitation wave-
lengths. The self-structured surface patterns formed on the star-
shaped azo compound films show close correlation with the type
of azo chormophores and light irradiation conditions. Upon light
irradiation with the same intensity (200 mW/cm2), efficient
surface pattern formation can be observed on the Tr-AZ-Cl, Tr-
35AZ-Cl, Tr-AZ-CN, and Tr-35AZ-CN films. Irradiation with
linearly and circularly polarized light shows significant differences
in the alignment manner of the pillarlike structures and their
saturated height. When the excitation wavelength is between the
λmax of the πꢀπ* transition and the band tail at the longer
wavelength side, the self-structured surface pattern can be more
efficiently induced by irradiation. In most cases, the 3,5-dimethyl
substitution on azo chromophores shows an effect of inhibiting
surface pattern formation for certain excitation wavelengths. The
exception to this rule could be attributed to the different path-
ways of photoinduced transꢀcis isomerization. The driving force
(22) Hiraoka, H.; Sendova, M. Appl. Phys. Lett. 1994, 64 (5), 563–565.
12675
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676

Langmuir
ARTICLE
(23) Ivanov, M.; Rochon, P. Appl. Phys. Lett. 2004, 84 (22), 4511–4513.
(24) Tsutsumi, N.; Fujihara, A. Appl. Phys. Lett. 2004, 85 (20),
4582–4584.
(25) Sipe, J. E.; Young, J. F.; Preston, J. S.; Vandriel, H. M. Phys. Rev.
B 1983, 27 (2), 1141–1154.
(26) Zhou, G. S.; Fauchet, P. N.; Siegman, A. E. Phys. Rev. B 1982, 26
(10), 5366–5382.
(27) Leblond, H.; Barille, R.; Ahmadi-Kandjani, S.; Nunzi, J. M.;
Ortyl, E.; Kucharski, S. J. Phys. B: At., Mol. Opt. Phys. 2009, 42, 205401.
(28) (a) Ahmadi-Kandjani, S.; Barille, R.; Dabos-Seignon, S.; Nunzi,
J. M.; Ortyl, E.; Kucharski, S. Opt. Lett. 2005, 30 (15), 1986–1988.
(b) Ahmadi-Kandjani, S.; Barille, R.; Dabos-Seignon, S.; Nunzi, J. M.;
Ortyl, E.; Kucharski, S. Mol. Cryst. Liq. Cryst. 2006, 446, 99–109.
(29) (a) Shirota, Y. J. Mater. Chem. 2000, 10, 1–25. (b) Shirota, Y.
J. Mater. Chem. 2005, 15, 75–93.
(30) (a) Thelakkat, M. Macromol. Mater. Eng. 2002, 287, 442–461.
€
(b) Ishow, E.; Bellaiche, C.; Bouteiller, L.; Nakatani, K.; Delaire, J. A.
J. Am. Chem. Soc. 2003, 125, 15744–15745.
(31) (a) Nakano, H.; Takahashi, T.; Kadota, T.; Shirota, Y. Adv.
Mater. 2002, 14, 1157–1160. (b) Nakano, H.; Tanino, T.; Takahashi, T.;
Ando, H.; Shirota, Y. J. Mater. Chem. 2008, 18, 242–246. (c) Ando, H.;
Tanino, T.; Nakano, H.; Shirota, Y. Mater. Chem. Phys. 2009, 113,
376–381.
(32) (a) Kim, M. J.; Seo, E. M.; Vak, D.; Kim, D. Y. Chem. Mater.
2003, 15, 4021–4027. (b) Chun, C.; Kim, M. J.; Vak, D.; Kim, D. Y.
J. Mater. Chem. 2003, 13, 2904–2909.
(33) Ishow, E.; Lebon, B.; He, Y. N.; Wang, X. G.; Bouteiller, L.;
Galmiche, L.; Nakatani, K. Chem. Mater. 2006, 18, 1261–1267.
(34) He, Y. N.; Gu, X. Y.; Guo, M. C.; Wang, X. G. Opt. Mater. 2008,
31, 18–27.
(35) Ishow, E.; Camacho-Aguilera, R.; Guꢀerin, J.; Brosseau, A.;
Nakatani, K. Adv. Funct. Mater. 2009, 19, 796–804.
(36) Yin, J. J.; Ye, G.; Wang, X. G. Langmuir 2010, 26 (9), 6755–6761.
(37) He, Y. N.; Wang, X. G.; Zhou, Q. X. Polymer 2002, 43, 7324–
7333.
(38) He, Y. N.; Yin, J. J.; Che, P. C.; Wang, X. G. Eur. Polym. J. 2006,
42, 292–301.
(39) Kim, M. J.; Lee, J. D.; Chun, C.; Kim, D. Y.; Higuchi, S.;
Nakayama, T. Macromol. Chem. Phys. 2007, 208, 1753–1763.
(40) Rau, H. In Photochemistry and Photophysics; Rabek, J. F., Ed.;
CRC pPress: Boca Raton, FL, 1990; Vol. II, Chapter 4.
(41) Brown, D.; Natansohn, A.; Rochon, P. Macromolecules 1995,
28, 6116–6123.
12676
dx.doi.org/10.1021/la2027253 |Langmuir 2011, 27, 12666–12676